NPU IP Core for Automotive

Overview

Whether deployed in-cabin for driver distraction or in the advanced driver assistance system (ADAS) stack for object recognition and point cloud processing, AI forms the backbone of the future of safer, smarter cars.

AI Demands Higher Performance

Auto makers are adding more AI, including advanced LLM and multimodal capabilities, as they enable safety and usability use cases such as autonomous driving/ADAS, driver attention monitoring, passenger detection, and infotainment. While local inference is essential for all safety-critical systems, LLMs may be 20 to 50X larger than more traditional AI networks. Automotive inference capabilities on today's processors are already limited, and automakers are looking for alternative, more efficient architectures for next generation systems.

Ideal Processing Architecture

Origin Evolution™ for Automotive offers out-of-the-box compatibility with popular LLM and CNN networks. Attention-based processing optimization and advanced memory management ensure optimal AI performance across a variety of today’s standard and emerging neural networks. Featuring a hardware and software co-designed architecture, Origin Evolution for Automotive scales to 96 TFLOPS in a single core, with multi-core performance to PetaFLOPs.

Innovative Architecture

Origin Evolution uses Expedera’s unique packet-based architecture to achieve unprecedented NPU efficiency. Packets, which are contiguous fragments of neural networks, are an ideal way to overcome the hurdle of large memory movements and differing network layer sizes, which are exacerbated by LLMs. Packets are routed through discrete processing blocks, including Feed Forward, Attention, and Vector, which accommodate the varying operations, data types, and precisions required when running different LLM and CNN networks. Origin Evolution includes a high-speed external memory streaming interface that is compatible with the latest DRAM and HBM standards.

Specifications

Compute Capacity	up to 48 FP16 MACs
Multi-tasking	Run Simultaneous Jobs
Example Networks Supported	Llama2, Llama3, ChatGLM, DeepSeek, Mistral, Qwen, MiniCPM, Yolo, MobileNet, and many others, including proprietary/black box networks
Example Performance	261 tokens per second, DeepSeek v3 token generation, 64 TFLOPS engine, batch size of 512, 256 GB/s external peak bandwith, 4.391W peak power consumption. Specified in TSMC 7nm, 1 GHz system clock, no sparsity/compression/pruning applied (though supported)
Layer Support	Standard NN functions, including Transformers, Conv, Deconv, FC, Activations, Reshape, Concat, Elementwise, Pooling, Softmax, others. Support for custom operators.
Data types	FP16/FP32/INT4/INT8/INT10/INT12/INT16 Activations/Weights
Quantization	Software toolchain supports Expedera, customer-supplied, or third-party quantization. Mixed precision supported.
Latency	Deterministic performance guarantees, no back pressure
Frameworks	Hugging Face, Llama.cpp, PyTorch, TVM, ONNX. Tensor Flow and others supported
Safety	ASIL-B readiness, ISO 9001:2015

Key Features

96 TFLOPS performance
Support for standard, custom, and proprietary neural networks
Readily customized for specific use cases and deployment needs
Full software stack provided, including compiler, estimator, scheduler, and quantizer
Runs LLM, CNN and other network types
Delivered as Soft IP (RTL) or GDS

Benefits

Choose the Features You Need: Customization brings many advantages, including increased performance, lower latency, reduced power consumption, and eliminating dark silicon waste. Expedera works with automotive customers to understand and optimize to their use case(s), PPA goals, and deployment needs during their design stage. Using this information, we configure Origin Evolution to create a customized solution that perfectly fits the application.
Reducing Memory Bandwidth: Origin Evolution's packet-architecture reduces memory requirements of popular LLMs like Llama 3.2 and Qwen1 by as much as 79%, saving system power and offering a much better utilized processor.
Efficient Resource Utilization: Origin Evolution for Automotive scales to 96 TFLOPS in a single core, eliminating the memory sharing, security, and area penalty issues faced by lower-performing, tiled AI accelerator engines. Origin Evolution NPUs achieve sustained utilization averaging 80%—compared to the 20-40% industry norm—avoiding dark silicon waste.
Full Software Stack: Origin Evolution employs an easy-to-use software stack that allows the importing of trained networks from popular representations such as Hugging Face, Llama.cpp, PyTorch, TVM, ONNX, TensorFlow, and others, while providing various quantization options, automatic completion, compilation, estimator and profiling tools. It also supports multi-job APIs.
LLM, CNN, and other Network Support: Origin Evolution offers out-of-the-box support for 100+ popular neural networks, including Llama2, Llama3, ChatGLM, DeepSeek, Mistral, Qwen, MiniCPM, Yolo, MobileNet, and many others.

Block Diagram

Technical Specifications

Request Info